CN114399654A - Method for identifying and alarming invasion target of power transmission line channel - Google Patents
Method for identifying and alarming invasion target of power transmission line channel Download PDFInfo
- Publication number
- CN114399654A CN114399654A CN202111332387.7A CN202111332387A CN114399654A CN 114399654 A CN114399654 A CN 114399654A CN 202111332387 A CN202111332387 A CN 202111332387A CN 114399654 A CN114399654 A CN 114399654A
- Authority
- CN
- China
- Prior art keywords
- transmission line
- power transmission
- identifying
- alarming
- domain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005540 biological transmission Effects 0.000 title claims abstract description 88
- 238000000034 method Methods 0.000 title claims abstract description 28
- 230000009545 invasion Effects 0.000 title description 4
- 238000000605 extraction Methods 0.000 claims abstract description 19
- 238000005259 measurement Methods 0.000 claims abstract description 4
- 230000003044 adaptive effect Effects 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 9
- 238000013135 deep learning Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000000691 measurement method Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 abstract description 5
- 238000012423 maintenance Methods 0.000 abstract description 4
- 230000008569 process Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20228—Disparity calculation for image-based rendering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a method for identifying and alarming an invading target of a power transmission line channel, which is used for identifying and alarming invading foreign matters in the power transmission line channel and comprises the following steps: step 1: acquiring an image of a transmission line channel by using a binocular measurement system; step 2: carrying out target identification on the image so as to identify the power transmission line and the invading foreign matter in the image; and step 3: carrying out feature extraction and feature matching on the power transmission line and the invading foreign matters in the image; and 4, step 4: performing three-dimensional reconstruction on the power transmission line and the invading foreign matter based on the results of the feature extraction and the feature matching; and 5: and calculating the distance between the invading foreign matter and the power transmission line based on the three-dimensional reconstruction result, judging whether the distance is less than a preset safety distance, and if so, giving an alarm. The invention is simple to implement and has strong implementability; the intelligent monitoring, identification and alarm in the whole process can be realized, and the manual operation and maintenance burden is greatly reduced.
Description
Technical Field
The invention belongs to the field of identification and alarm of foreign matters invading a power transmission line channel, and particularly relates to a method for identifying and alarming an invading target of a power transmission line channel.
Background
The power transmission line channel is a strip-shaped region below a region of a predetermined width extending from the high-voltage overhead power transmission line to both sides. Access to personnel and related agricultural or production activities are severely restricted in the ground portion of the area; meanwhile, in order to ensure personal and property safety, necessary clean space is required to be reserved around the high-voltage wire, so that the strength of the static induction electric field of the high-voltage wire is weakened to the extent of not endangering personal safety. For example, a single-circuit 500 kV ultrahigh voltage transmission line has the requirement that the height of a tower is more than 30 meters and the width of a line channel is about 45 meters.
Under normal conditions, people, vehicles and the like can avoid power transmission facilities consciously; however, when necessary construction and electric power facility maintenance are carried out in the electric power transmission line channel, due to the fact that training and supervision are not sufficient, the situation that part of ultrahigh engineering vehicles break into the channel by mistake still exists, and casualty accidents are caused accordingly. Therefore, it is necessary to arrange related intelligent monitoring, identifying and warning devices in the power transmission line channel so as to remind intruders in time and guarantee the safety of lives and properties.
Disclosure of Invention
The invention aims to provide a method for identifying and alarming the invading target of the power transmission line channel, which can solve the problem of identifying the invading target in the power transmission line channel so as to ensure the safety of production and life.
In order to achieve the purpose, the invention adopts the technical scheme that:
a method for identifying and alarming an invading target of a power transmission line channel is used for identifying and alarming invading foreign matters in the power transmission line channel, and comprises the following steps:
step 1: acquiring an image of the transmission line channel by using a binocular measurement system;
step 2: carrying out target identification on the image so as to identify the power transmission line and the invading foreign matter in the image;
and step 3: carrying out feature extraction and feature matching on the power transmission line and the invading foreign matter in the image;
and 4, step 4: performing three-dimensional reconstruction on the power transmission line and the invasive foreign matter based on the results of feature extraction and feature matching;
and 5: and calculating the distance between the invading foreign matter and the power transmission line based on the three-dimensional reconstruction result, judging whether the distance is less than a preset safety distance, and if so, giving an alarm.
In the step 2, the image is subjected to target recognition by using a deep learning framework, wherein the deep learning framework comprises a YOLO v4 detector framework trained by a domain adaptive network.
In the step 2, the YOLO v4 detector framework extracts three features of different scales, the domain adaptive network takes the three features of different scales extracted by the YOLO v4 detector framework as input, the three features are respectively accessed to the domain adaptive network through corresponding three gradient reversion layers, each feature respectively passes through two convolutional layer prediction domain class probabilities in the domain adaptive network, and then passes through a domain classifier layer to calculate a domain classification loss, and a calculation function of the domain classification loss is as follows:
wherein t isiIs the ground truth domain label, t, of the ith training imageiWhen 1, it is the source domain; t is tiWhen the value is 0, the target domain is obtained;
the feature map at position (x, y) for the ith training image.
The YOLO v4 detector framework passes through a minimization of a loss function
To optimize its backbone network, wherein,
to detect the loss function, λ is a negative scalar of the gradient inversion layer.
And 3, extracting the edge of the power transmission line and the characteristic points of the invaded foreign matters, and matching the same characteristic points in the two images acquired by the binocular measuring system.
In the step 3, the edges of the power transmission lines are extracted by using a Hough transformation straight line detection method, and the feature points of the invading foreign bodies are extracted by using an SIFT feature extraction algorithm.
And eliminating mismatching points in the two images by using a RANSAC algorithm during feature matching.
In the step 4, a binocular stereo vision three-dimensional measurement method is used for obtaining the three-dimensional space coordinates of the feature points extracted by feature extraction, the transmission line is regarded as a straight line formed by a plurality of feature points, and a corresponding straight line equation is obtained, so that the transmission line and the invasive foreign matters are subjected to three-dimensional reconstruction.
And 5, calculating the distance from the characteristic point on the invaded foreign matter to the straight line of the characteristic point of the power transmission line by using a distance calculation formula from the middle point to the straight line in space.
And in the step 5, judging whether the minimum value of the distance from the characteristic point on the invading foreign matter to the straight line of the characteristic point of the power transmission line is less than a preset safety distance.
Due to the application of the technical scheme, compared with the prior art, the invention has the following advantages: the invention is simple to implement and has strong implementability; the intelligent monitoring, identification and alarm in the whole process can be realized, and the manual operation and maintenance burden is greatly reduced.
Drawings
FIG. 1 is a flow chart of the method for identifying and alarming the intrusion object of the power transmission line channel.
Fig. 2 is a structural diagram of a deep learning framework adopted in the method for identifying and alarming an intrusion object of a power transmission line channel of the invention.
Fig. 3 is a schematic diagram of the invasion of foreign matters in the transmission line channel.
Detailed Description
The invention will be further described with reference to examples of embodiments shown in the drawings to which the invention is attached.
The first embodiment is as follows: as shown in fig. 1, a method for identifying and alarming an intrusion object in a power transmission line channel for identifying and alarming an intrusion foreign object in the power transmission line channel includes the following steps:
step 1: image acquisition
And obtaining an image of the transmission line channel by using a binocular measuring system.
A binocular camera is built near a power line tower pole to form a binocular measuring system, the binocular camera is used for shooting and obtaining images in the range of a channel of a power transmission line, the binocular camera of the binocular measuring system can shoot a target area from different shooting positions at the same moment, and two images of the target area can be obtained at the same time. The adopted binocular measuring system can adopt the existing binocular camera products on the market and can also be built by two cameras.
Step 2: object recognition
And carrying out target identification on the image so as to identify the power transmission line and the invading foreign matter in the image.
In this step, based on the two obtained images, the power line and the invading foreign matter (for example, an engineering vehicle invading into the power line channel, etc.) in the images are subjected to target recognition by using the depth learning framework. The deep learning framework employs a multi-scale adaptive deep learning framework that includes a domain-adaptive network-trained YOLO v4 detector framework, which generates domain-invariant features using multiple domain-adaptive paths and corresponding domain classifiers on different scales of a YOLO v4 detector.
The deep learning framework shown in fig. 2 is added with a domain adaptive network on the basis of the YOLO v4 detector framework. The YOLO v4 detector framework extracts three features of different scales, which include three main parts, namely a backbone network, a bottleneck module, and a prediction module. The backbone network is responsible for extracting multilayer features of the input image under different scales, wherein the multilayer features comprise features F1, F2 and F3; the bottleneck module gathers the extracted features (F1, F2 and F3) of three different scales in the backbone network through an upsampling layer and inputs the features into the prediction module; the prediction module calculates boundary boxes around the target to be recognized and the class probability associated with each boundary box, thereby realizing the detection and recognition of the target in the image. The domain adaptive network is only appended to YOLO v4 at training time to learn domain invariant features; and after the training is finished, the original YOLO v4 detector (without adding a domain adaptive network) is still used for target detection and identification.
The domain adaptive network takes three features F1, F2 and F3 of different scales extracted by a YOLO v4 detector framework as input, and the three features are respectively accessed to the domain adaptive network through corresponding three gradient reversion layers. On each scale, after the domain class probability of each feature is predicted by two convolutional layers in the domain adaptive network, the domain classification loss is calculated by a domain classifier layer. Specifically, two convolutional layers are respectively connected behind the gradient inversion layer, wherein the first convolutional layer reduces the characteristic channel by half, that is, the first convolutional layer corresponding to the characteristic F1 is 128 channels, the first convolutional layer corresponding to the characteristic F2 is 256 channels, the first convolutional layer corresponding to the characteristic F3 is 512 channels, and the second convolutional layers corresponding to the three characteristics are all 1 channel and are used for predicting the domain class probability; and finally, connecting a domain classifier layer behind the convolution layer, and calculating domain classification loss by using the domain classifier layer.
The computational function of the domain classification penalty is:
wherein t isiIs the ground truth domain label, t, of the ith training imageiWhen 1, it is the source domain; t is tiWhen the value is 0, the target domain is obtained;
the feature map at position (x, y) for the ith training image. The source domain and the target domain are distinguished by minimizing the loss function described above.
In the YOLO v4 Detector framework, by minimizing the detection loss function
To perform the optimization. And on the other hand, classifying the loss function for the domain
Then, by minimizing
Optimizing domain adaptive networks, maximization
Optimizing the backbone network, i.e. taking a pairAn anti-learning strategy; thus the final YOLO v4 detector framework passes the minimization of the loss function
To optimize its backbone network, wherein,
for detecting the loss function, λ is a negative scalar of the gradient inversion layer, which is used to balance the loss function
Sum domain classification penalty function
Through the improved YOLO v4 detector framework, the detection performance of a target domain can be improved, and the method is suitable for detecting and identifying the foreign object in the complex environment where the power transmission line channel is located.
And step 3: feature extraction
The method comprises the steps of carrying out feature extraction and feature matching on a power transmission line and an invading foreign matter in an image, specifically extracting the edge of the power transmission line and the feature point of the invading foreign matter, and matching the same feature point in two images acquired by a binocular measurement system. Because the binocular camera simultaneously acquires the images at the two shooting positions, the same characteristic point in the two images needs to be matched after the characteristic points of the target object in the two images are respectively extracted.
3.1 edge extraction of Power lines
The transmission line can be approximately seen as a straight line and is formed by connecting a plurality of characteristic points in the image. In the step, the edge of the power transmission line is extracted by using a Hough transformation straight line detection method. Firstly, extracting the image edge of an ROI (region of interest) of a power transmission line by using a Canny edge extraction algorithm, and smoothing the extracted edge contour by using a Gaussian function so as to have stronger robustness to image noise; and then, extracting the edge of the power line image through Hough transformation.
For the edges of the power transmission line extracted from the two images, the epipolar geometric constraint is utilized, and the RANSAC algorithm is combined to remove mismatching points in the two images, so that the precise registration of the characteristic points between the two images is realized.
3.2 feature point extraction of invading foreign bodies
In order to reduce the influence of external environment light on feature extraction, feature points in an ROI (region of interest) of an invaded foreign matter are extracted by using an SIFT feature extraction algorithm, and then mismatching points in two images are removed by using an RANSAC algorithm, so that the SIFT feature points between the two images are accurately registered.
And 4, step 4: three-dimensional reconstruction
And performing three-dimensional reconstruction on the power transmission line and the invasive foreign matters based on the results of the feature extraction and the feature matching.
The three-dimensional space coordinates of the feature points extracted by feature extraction are obtained by using a binocular stereo vision three-dimensional measurement method, the power transmission line is regarded as a straight line formed by a plurality of feature points, and a corresponding straight line equation is obtained, so that three-dimensional reconstruction is carried out on the power transmission line and the invading foreign matters.
The binocular stereo vision three-dimensional measurement method is based on the parallax principle, and assumes that two lenses of a binocular camera are respectively a left lens and a right lens, and images shot by the two lenses are on the same plane; for the same feature point P in space, let its coordinate in the actual three-dimensional space be (x)c，yc，zc) And the coordinate P of the feature point P on the left-shot imageL＝(XL，YL) Coordinate P on right-hand shot imageR＝(XR，YR) And then:
wherein f is the focal length of the camera lens, and B is the baseline distance, i.e. the distance between the projection centers of the two lenses.
According to the parallax D ═ XL-XRThen, the three-dimensional coordinates of the feature point P can be represented by the formula:
and (4) calculating.
By the method, the three-dimensional space coordinates of the characteristic points can be obtained; for the power transmission line, a linear equation ax + by + cz + d of the power transmission line in the actual three-dimensional space can be obtained by determining the space coordinates of the characteristic points; for the invasion foreign matter, only the judgment z is neededcDetermines the feature point at its highest position.
And 5: distance calculation and determination
And calculating the distance between the invading foreign matter and the power transmission line based on the three-dimensional reconstruction result, judging whether the distance is less than a preset safety distance, and if so, giving an alarm.
After the position information of the power transmission line and the invading foreign body in the three-dimensional space is known, the distance between the characteristic point on the invading foreign body and the straight line where the characteristic point of the power transmission line is located is calculated by utilizing a distance calculation formula from the middle point to the straight line in the space.
As shown in FIG. 3, the transmission lines are regarded as straight lines, and the equation of the straight line of the ith transmission line is aix+bix+ciz+diThe coordinate of the feature point at the highest position of the invading foreign body is set as (x)0，y0，z0) And then the distance between the invading foreign matter and the ith transmission line is as follows:
the distance of invading the foreign matter apart from every power transmission line is solved in proper order, and whether it is less than predetermined safe distance is judged to the minimum of the distance between the characteristic point place straight line of each power transmission line to the characteristic point on utilizing the invading foreign matter, obtains wherein minimum distance through the comparison promptly:
Dmin＝min(Di)
finally, judging the minimum distance DminAnd judging whether the potential safety hazard exists in the power transmission line channel or not and sending corresponding alarm information if the potential safety hazard exists in the power transmission line channel.
The invention provides a method for identifying and warning an invading target of a power transmission line channel, which comprises the steps of erecting a binocular camera in the monitoring range of the power transmission line channel, identifying a target object in an image by an image identification technology, reconstructing position information of the target object in a three-dimensional space according to images at two different shooting positions acquired by the binocular camera based on a binocular stereo vision three-dimensional measurement method, and finally calculating the distance from an invading foreign body to the power transmission line so as to accurately judge whether the invading risk of the foreign body exists or not according to the distance. The beneficial effects are that: (1) the required hardware equipment is relatively simple, the main hardware equipment is only a binocular camera, and the binocular camera belongs to the existing mature product, so the identification warning method has strong feasibility of implementation; (2) the intelligent monitoring, identification and alarm in the whole process can be realized, and the manual operation and maintenance burden is greatly reduced.
The above embodiments are merely illustrative of the technical ideas and features of the present invention, and the purpose thereof is to enable those skilled in the art to understand the contents of the present invention and implement the present invention, and not to limit the protection scope of the present invention. All equivalent changes and modifications made according to the spirit of the present invention should be covered within the protection scope of the present invention.
Claims (10)
1. A method for identifying and alarming an invading object of a transmission line channel is used for identifying and alarming invading foreign matters in the transmission line channel and is characterized in that: the method for identifying and alarming the intrusion target of the power transmission line channel comprises the following steps:
step 1: acquiring an image of the transmission line channel by using a binocular measurement system;
step 2: carrying out target identification on the image so as to identify the power transmission line and the invading foreign matter in the image;
and step 3: carrying out feature extraction and feature matching on the power transmission line and the invading foreign matter in the image;
and 4, step 4: performing three-dimensional reconstruction on the power transmission line and the invasive foreign matter based on the results of feature extraction and feature matching;
and 5: and calculating the distance between the invading foreign matter and the power transmission line based on the three-dimensional reconstruction result, judging whether the distance is less than a preset safety distance, and if so, giving an alarm.
2. The method for identifying and alarming targets invaded by power transmission line channels as claimed in claim 1, characterized in that: in the step 2, the image is subjected to target recognition by using a deep learning framework, wherein the deep learning framework comprises a YOLO v4 detector framework trained by a domain adaptive network.
3. The method for identifying and alarming targets invaded by power transmission line channels as claimed in claim 2, characterized in that: in the step 2, the YOLO v4 detector framework extracts three features of different scales, the domain adaptive network takes the three features of different scales extracted by the YOLO v4 detector framework as input, the three features are respectively accessed to the domain adaptive network through corresponding three gradient reversion layers, each feature respectively passes through two convolutional layer prediction domain class probabilities in the domain adaptive network, and then passes through a domain classifier layer to calculate a domain classification loss, and a calculation function of the domain classification loss is as follows:
wherein t isiIs the ground truth domain label, t, of the ith training imageiWhen 1, it is the source domain; t is tiWhen the value is 0, the target domain is obtained;
the feature map at position (x, y) for the ith training image.
4. A power transmission line according to claim 3The method for identifying and alarming the road intrusion object is characterized by comprising the following steps: the YOLO v4 detector framework passes through a minimization of a loss function
To optimize its backbone network, wherein,
to detect the loss function, λ is a negative scalar of the gradient inversion layer.
5. The method for identifying and alarming targets invaded by power transmission line channels as claimed in claim 1, characterized in that: and 3, extracting the edge of the power transmission line and the characteristic points of the invaded foreign matters, and matching the same characteristic points in the two images acquired by the binocular measuring system.
6. The method for identifying and alarming targets invaded by power transmission line channels as claimed in claim 5, characterized in that: in the step 3, the edges of the power transmission lines are extracted by using a Hough transformation straight line detection method, and the feature points of the invading foreign bodies are extracted by using an SIFT feature extraction algorithm.
7. The method for identifying and alarming targets invaded by power transmission line channels as claimed in claim 6, characterized in that: and eliminating mismatching points in the two images by using a RANSAC algorithm during feature matching.
8. The method for identifying and alarming targets invaded by power transmission line channels as claimed in claim 1, characterized in that: in the step 4, a binocular stereo vision three-dimensional measurement method is used for obtaining the three-dimensional space coordinates of the feature points extracted by feature extraction, the transmission line is regarded as a straight line formed by a plurality of feature points, and a corresponding straight line equation is obtained, so that the transmission line and the invasive foreign matters are subjected to three-dimensional reconstruction.
9. The method for identifying and alarming targets invaded by power transmission line channels as claimed in claim 1, characterized in that: and 5, calculating the distance from the characteristic point on the invaded foreign matter to the straight line of the characteristic point of the power transmission line by using a distance calculation formula from the middle point to the straight line in space.
10. The method for identifying and alarming targets invaded by power transmission line channels as claimed in claim 1, characterized in that: and in the step 5, judging whether the minimum value of the distance from the characteristic point on the invading foreign matter to the straight line of the characteristic point of the power transmission line is less than a preset safety distance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111332387.7A CN114399654A (en) | 2021-11-11 | 2021-11-11 | Method for identifying and alarming invasion target of power transmission line channel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111332387.7A CN114399654A (en) | 2021-11-11 | 2021-11-11 | Method for identifying and alarming invasion target of power transmission line channel |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114399654A true CN114399654A (en) | 2022-04-26 |
Family
ID=81225590
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111332387.7A Pending CN114399654A (en) | 2021-11-11 | 2021-11-11 | Method for identifying and alarming invasion target of power transmission line channel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114399654A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117058518A (en) * | 2023-08-03 | 2023-11-14 | 南方电网数字电网研究院有限公司 | Deep learning target detection method and device based on YOLO improvement and computer equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109801302A (en) * | 2018-12-14 | 2019-05-24 | 华南理工大学 | A kind of ultra-high-tension power transmission line foreign matter detecting method based on binocular vision |
| CN112434586A (en) * | 2020-11-16 | 2021-03-02 | 中山大学 | Multi-complex scene target detection method based on domain adaptive learning |
| US20210064912A1 (en) * | 2018-01-26 | 2021-03-04 | Sony Interactive Entertainment Inc. | Learning apparatus, system for generating captured image classification apparatus, apparatus for generating captured image classification apparatus, learning method, and program |
| CN113076994A (en) * | 2021-03-31 | 2021-07-06 | 南京邮电大学 | Open-set domain self-adaptive image classification method and system |
| CN113486819A (en) * | 2021-07-09 | 2021-10-08 | 广西民族大学 | Ship target detection method based on YOLOv4 algorithm |
-
2021
- 2021-11-11 CN CN202111332387.7A patent/CN114399654A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210064912A1 (en) * | 2018-01-26 | 2021-03-04 | Sony Interactive Entertainment Inc. | Learning apparatus, system for generating captured image classification apparatus, apparatus for generating captured image classification apparatus, learning method, and program |
| CN109801302A (en) * | 2018-12-14 | 2019-05-24 | 华南理工大学 | A kind of ultra-high-tension power transmission line foreign matter detecting method based on binocular vision |
| CN112434586A (en) * | 2020-11-16 | 2021-03-02 | 中山大学 | Multi-complex scene target detection method based on domain adaptive learning |
| CN113076994A (en) * | 2021-03-31 | 2021-07-06 | 南京邮电大学 | Open-set domain self-adaptive image classification method and system |
| CN113486819A (en) * | 2021-07-09 | 2021-10-08 | 广西民族大学 | Ship target detection method based on YOLOv4 algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 肖贵平等: "《十二五普通高等教育本科国家级规划教材 交通安全工程 第3版》", 31 March 2021, 中国铁道出版社, pages: 274 - 276 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117058518A (en) * | 2023-08-03 | 2023-11-14 | 南方电网数字电网研究院有限公司 | Deep learning target detection method and device based on YOLO improvement and computer equipment |
| CN117058518B (en) * | 2023-08-03 | 2024-05-03 | 南方电网数字电网研究院有限公司 | Deep learning target detection method and device based on YOLO improvement and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106407931B (en) | A kind of depth convolutional neural networks moving vehicle detection method | |
| CN110532889B (en) | Track foreign matter detection method based on rotor unmanned aerial vehicle and YOLOv3 | |
| CN104574376B (en) | Avoiding collision based on binocular vision and laser radar joint verification in hustle traffic | |
| Qu et al. | Moving vehicle detection with convolutional networks in UAV videos | |
| CN111611874B (en) | Face mask wearing detection method based on ResNet and Canny | |
| Akarsu et al. | A fast and adaptive road defect detection approach using computer vision with real time implementation | |
| CN106548147A (en) | A kind of quick noise robustness image foreign matter detection method and TEDS systems | |
| CN110660065B (en) | Infrared fault detection and identification algorithm | |
| CN107480646B (en) | Binocular vision-based vehicle-mounted video abnormal motion detection method | |
| CN112329623A (en) | Early warning method for visibility detection and visibility safety grade division in foggy days | |
| CN113409252B (en) | Obstacle detection method for overhead transmission line inspection robot | |
| CN111667655A (en) | Infrared image-based high-speed railway safety area intrusion alarm device and method | |
| Vancea et al. | Vehicle taillight detection and tracking using deep learning and thresholding for candidate generation | |
| Šilar et al. | The obstacle detection on the railway crossing based on optical flow and clustering | |
| CN108189757A (en) | A kind of driving safety prompt system | |
| CN115100741B (en) | Point cloud pedestrian distance risk detection method, system, equipment and medium | |
| CN114399654A (en) | Method for identifying and alarming invasion target of power transmission line channel | |
| Jiang et al. | Foreign object recognition technology for port transportation channel based on automatic image recognition | |
| Yi et al. | Safety belt wearing detection algorithm based on human joint points | |
| CN112508893B (en) | Method and system for detecting tiny foreign matters between double rails of railway based on machine vision | |
| CN114842340A (en) | Robot binocular stereoscopic vision obstacle sensing method and system | |
| CN110826468B (en) | Driving-assisted vehicle detection distance measurement method based on lane line detection | |
| CN115171006B (en) | Detection method for automatically identifying person entering electric power dangerous area based on deep learning | |
| CN115100618B (en) | Multi-source heterogeneous perception information multi-level fusion characterization and target identification method | |
| Qin et al. | An improved deep learning algorithm for obstacle detection in complex rail transit environments |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |